Re: [PATCH] Linux: Do not skip d_ino == 0 entries in readdir, readdir64 (bug 12165)

[Florian Weimer <fweimer@redhat.com>]

* Andreas Schwab:

> On Sep 19 2022, Florian Weimer wrote:
>
>> * Andreas Schwab:
>>
>>> On Sep 19 2022, Florian Weimer via Libc-alpha wrote:
>>>
>>>> -	  if (bytes <= 0)
>>>> -	    {
>>>> -	      /* On some systems getdents fails with ENOENT when the
>>>
>>> Shouldn't that be "some file systems"?
>>>
>>>> -		 open directory has been rmdir'd already.  POSIX.1
>>>> -		 requires that we treat this condition like normal EOF.  */
>>>> -	      if (bytes < 0 && errno == ENOENT)
>>>> -		bytes = 0;
>>>> -
>>>> -	      /* Don't modifiy errno when reaching EOF.  */
>>>> -	      if (bytes == 0)
>>>> -		__set_errno (saved_errno);
>>>> -	      dp = NULL;
>>>> -	      break;
>>>> -	    }
>>>> -	  dirp->size = (size_t) bytes;
>>>> -
>>>> -	  /* Reset the offset into the buffer.  */
>>>> -	  dirp->offset = 0;
>>>> +	  /* Linux fails with ENOENT if IS_DEADDIR is true for the
>>>
>>> What is IS_DEADDIR?
>>
>> It's an internal Linux construct.  Isn't this clear from the context?
>
> I'd rather keep the original comment with the missing word added.  I
> find it easier to understand without the reference to kernel internals.

What about this?

	  /* Linux may fail with ENOENT on some file systems if the
	     directory inode is marked as dead (deleted).  POSIX
	     treats this as a regular end-of-directory condition, so
	     do not set errno in that case, to indicate success.  */

Thanks,
Florian